© 2004 American Thoracic Society
Treatment of Multidrug-resistant TuberculosisShould You Try This at Home?University of Texas Health Center Tyler, Texas In 1993, Goble and coworkers published treatment outcomes for 171 patients with multidrug-resistant tuberculosis (TB) from one institution over a 10-year period (1). The initial and overall favorable response rates were 65 and 56%, respectively, with 37% mortality, approximately half attributed to TB. A companion study from the same institution demonstrated that physician-related treatment errors led to the development of multidrug-resistant TB patients whose therapy was then associated with "extraordinary" costs (2). An editorial accompanying the article by Goble and coworkers proposed directly observed TB therapy as a mechanism for reducing treatment errors, thereby reducing the risk of creating new drug-resistant TB cases (1, 3). The editorial effectively argued that even though the initial costs of directly observed therapy were greater than unsupervised therapy, "we can't afford not to try it." The costs of directly observed therapy subsequently proved to be amply compensated for by reductions in both TB treatment failure and the attendant risks of acquired drug resistance (4). Directly observed therapy is now routinely recommended for treatment of TB in the U.S. The small and relatively stable number of new multidrug-resistant TB cases over the last 10 years is evidence of the effectiveness of this strategy for preventing new drug-resistant cases (5). In the U.S., however, there has been an important shift in TB epidemiology, with more than 50% of new TB cases and approximately 69% of the multidrug-resistant TB cases diagnosed in persons born outside the U.S.; patients who, presumably, acquired drug-resistant TB before entering the U.S. (5). We may have reached the limits in the U.S. for reducing multidrug-resistant TB through application of directly observed therapy. Further case reductions will require more effective treatment strategies for multidrug-resistant TB. In this issue of the Journal (pp. 1103–1109), Chan and coworkers (6) update treatment outcome for multidrug-resistant TB patients treated at their institution over a 14-year period. Compared with the previous study by Goble and coworkers (1), initial favorable response rates improved to 85%, long-term success rates to 75%, and tuberculosis death rates decreased to 12% (6). The most significant factor associated with improved outcome was TB-related surgery. Fluoroquinolone use had a statistically significant favorable effect on outcome for a subsegment of the population. Two critical elements are inextricably linked with the approach described by Chan and coworkers: clinical expertise and adequate funding (6). In an era of declining support for TB control, is this model of therapy for multidrug-resistant TB widely applicable? A contribution by surgery to successful treatment of multidrug-resistant TB is perhaps not unexpected, but surgery as the most important factor for favorable outcome is surprising. A recent review of several studies reporting treatment outcome for multidrug-resistant TB patients suggested a role for surgery (7), but not a dominant role as indicated by Chan and coworkers (6). The surgeons involved in this study are, unquestionably, the most experienced mycobacterial lung disease surgeons in the U.S. (6, 8). An important aspect of that experience is the ability to choose patients most likely to benefit from surgery. As evident in the study by Chan and coworkers, rigorous or widely accepted preoperative criteria are not available (6). It would, therefore, be difficult for surgeons inexperienced in mycobacterial lung disease surgery to assess the risk/benefit balance of surgery for individual patients. Mycobacterial lung disease surgery also requires special surgical skills and can be associated with significant and unpredictable surgical complications (8, 9). It is unlikely that acceptably low surgical morbidity and mortality could be maintained at centers that infrequently encounter these patients. Also surprising was the relatively modest contribution of fluoroquinolones, which were associated with improved outcome but for only a segment of the study population (6). Recent promising developments in TB drug therapy could boost that effect. The 8-methoxy-fluoroquinolones are more active against TB than older fluoroquinolones, and the oxazolidinone, linezolid, has excellent activity against TB (10–12). The cost of including these newer agents in multidrug-resistant TB regimens would be substantial, however, and it is unclear if local and state health departments would have the resources to supply these new drugs. An implied, but not explicitly credited, major contribution to both favorable outcome and expense of patient care in the study of Chan and coworkers was prolonged hospitalization (6). Hospitalization of contagious multidrug-resistant TB patients undeniably benefited the public health and greatly facilitated the aggressive treatment modalities. Decreased funding, increased costs, and limited clinical expertise are not an encouraging combination for supporting the widespread adoption of the treatment model described by Chan and coworkers (6). It is also fair to ask how many centers in the U.S. have expertise in mycobacterial lung disease surgery and how many centers could offer such comprehensive care to a largely uninsured population? A successful strategy to address both the financial demands and the demands for maintaining clinical expertise could be accomplished through concentration of effort in regional centers for treatment of multidrug-resistant TB patients, with pooling of resources from multiple sources including state and national public health authorities. Regional centers would also facilitate initiation of multidrug-resistant TB treatment trials. Important issues such as the role or necessity of surgery and the optimal number of drugs necessary for successful therapy could be addressed with concentration of adequate numbers of patients for study. A weakness of the studies by Goble and coworkers (1) and Chan and coworkers (6) was the lack of outpatient case management after patients were discharged, with many patients lost to follow-up. After an initial period of intense medical and surgical therapy, patients could be treated in their communities with ongoing support of local physicians by the regional centers. An outpatient case management component would greatly improve continuity and quality of care. Is there a less expensive model for treating multidrug-resistant TB patients with equal or better success? Mitnick and coworkers (13) recently reported the results of a community-based program from Peru with treatment success comparable to that reported by Chan and coworkers (6). The treatment approach by Mitnick and coworkers was still relatively expensive, and likely not applicable to resource-poor areas. The success of their program also depended on several features that would be part of the proposed regional centers, including excellent microbiologic support, clinical TB expertise, and aggressive case management. The study by Chan and coworkers (6) could have an impact on the direction of multidrug-resistant TB therapy in the U.S., but, like the study by Goble and coworkers (1), that would require a change in the priorities for funding TB therapy. An entirely outpatient approach to multidrug-resistant TB patients would not be acceptable in the U.S. for public health reasons. Lack of funding and clinical expertise mitigate against widespread hospitalization of multidrug-resistant TB patients across the U.S. Regional centers for management of multidrug-resistant TB cases offer several major advantages, including maintenance of cognitive and technical skills for care of these complicated patients, facilitation of collection and analysis of patient data, initiation of treatment trials with adequate numbers of patients, ongoing case management by experts, and perhaps most importantly, shared resources for the considerable financial burden of treating these patients. FOOTNOTES Conflict of Interest Statement: D.E.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
